Home Credit Default Risk Assessment

About Risk Assessment Problem

This problem relates to assessing risk of default on the part of a person seeking loan at Home Credit. Home Credit aims to empower those who have little or no credit history. Its website states:

Our services are simple, easy and fast. Our responsible lending model empowers underserved customers with little or no credit history to access financing, enabling them to borrow easily and safely, both online and offline.

Journey of a client into Home Credit services is in step-wise progression. First, the client begins by availing of Home Credit loan through a product purchase or what is called point-of-sale (POS) loans. Reliable customers can then adopt broader consumer credit products and ultimately we progress to providing fully fledged branch-based consumer lending.

In order to make sure this underserved population has a positive loan experience, Home Credit makes use of a variety of alternative data–including telco and transactional information–to predict their clients’ repayment abilities.

While Home Credit is currently using various statistical and machine learning methods to make these predictions, they’re challenging Kagglers to help them unlock the full potential of their data. Doing so will ensure that clients capable of repayment are not rejected and that loans are given with a principal, maturity, and repayment calendar that will empower their clients to be successful.

Data for the purpose has been released on Kaggle and is available at here.

This book has been written from the point of view of explaining how through simple aggregation operations over various features, new features can be generated and a good level of prediction made facilitating the decision to advance or not advance a loan.

We use the code of jsaguiar Aguiar, Kaggle Master available on Kaggle at this link. We have profusely commented the code, made some changes (so as to keep it simple for beginners in Machine Learning). Also we use some material from this highly upvoted notebook of Will Koehrsen, Kaggle Master.

This book is a work in progress.